MOSAIKS On Microsoft Azure Machine Learning Studio #211
Unanswered
CarloBroderick
asked this question in
Q&A
Replies: 1 comment 2 replies
-
Can you share
We do have some rate limits on the STAC API to keep the service health, but those wouldn't result in that error message. And authentication wouldn't help your rate limit on the STAC API (just the SAS API when you're signing items). |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello MPC Community,
First time, long time.
My name is Carlo Broderick, I am a grad student at UC Santa Barbara following up on a project done at the Bren School last year in which Cullen Molitor and three other grad students implemented the MOSAIKS process to predict Zambian distinct level maize yield.
The MOSAIKS process is helpfully outlined in one of MPCs tutorials. Our team originally attempted to use MPC for our MOSAIKS process; however, we found that it was difficult to reliably access MPC's PyTorch instance and extract features, likely because our process was more computationally intensive than last years.
After migrating over to Microsoft Machine Learning studio and setting up a VM in Western Europe my team is running into some difficulty implementing the MOSAIKS process on Azure. Namely, our process seems to error intermittently with a message:
"Service unavailable
Our services aren't available right now
We're working to restore all services as soon as possible. Please check back soon."
The nature of this error and the intermittency make me wonder if this is a symptom of our requests being throttled down and we are requesting too many STAC items. Would authentication help this error? Would I just need to submit my token along with the HTML requests in the STAC api?
Happy to give more information about the problem if that will help resolution.
Thanks for your support.
Love what MPC is all about.
Best,
CB
Beta Was this translation helpful? Give feedback.
All reactions